This prospective cross-sectional study was conducted to determine the accuracy of Goodsall's rule in predicting the location of the internal opening of an anal fistula based on the location of the external opening among Filipino patients and the clinical factors that significantly influenced the accuracy of Goodsall's rule. From January 15, 1997 to April 15, 1998, 102 adult patients were analyzed by age, gender, fistula type, previous fistula surgery, symptom duration, location of the external opening, type of fistulous tract and distance of external opening from the anal verge. Overall, in 79 patients (77.5%), the locations of the internal opening were accurately localized according to Goodsall's rule. Univariate analysis showed that only the location of the external opening (posterior vs. anterior) and the distance of the external opening from the anal verge ((3 cm vs.> 3 cm) demonstrated a significant correlation with Goodsall's rule. Multivariate logistic regression analysis further demonstrated that anteriorly located external openings and distance of the opening greater than 3 cm from the anal verge were unlikely to comply with Goodsall's rule (odds ratio=0.008). Only 2 of 14 patients (14.3%) with anterior external openings greater than 3 cm from the anal verge had fistulous tracts which followed this rule. Thus, surgeons are advised to exercise caution during the surgical evaluation and treatment of anal fistula with these characteristics.